Product Compare
NVIDIA GRID RTX T10 2 vs NVIDIA GB10
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2020 | Oct 2025 |
| Generation | GRID | Server Blackwell |
| Type | Professional | Desktop |
| Bus Interface | PCIe 3.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 1065 MHz | 1665 MHz |
| Boost Clock | 1395 MHz | 2418 MHz |
| Memory Clock | 1750 MHz | 1067 MHz |
| Memory | ||
| Memory Size | 2GB | 128GB |
| Memory Type | GDDR6 | LPDDR5X |
| Memory Bus | 384bit | 256bit |
| Bandwidth | 672.0GB/s | 273.2GB/s |
| Render Config | ||
| SM Count | 56 | 48 |
| Shading Units | 3584 | 6144 |
| TMUs | 224 | 384 |
| ROPs | 64 | 48 |
| Tensor Cores | 448 | 384 |
| RT Cores | 56 | 48 |
| L1 Cache | 64 KB (per SM) | 256 KB (per SM) |
| L2 Cache | 6 MB | 50 MB |
| Theoretical Performance | ||
| Pixel Rate | 89.28 GPixel/s | 116.1 GPixel/s |
| Texture Rate | 312.5 GTexel/s | 928.5 GTexel/s |
| FP16 (half) | 20.00 TFLOPS | 29.71 TFLOPS |
| FP32 (float) | 9.999 TFLOPS | 29.71 TFLOPS |
| FP64 (double) | 312.5 GFLOPS | 464.3 GFLOPS |
| Graphics Processor | ||
| GPU Name | TU102 | GB20B |
| GPU Variant | TU102-875-A1 | |
| Architecture | Turing | Blackwell 2.0 |
| Foundry | TSMC | TSMC |
| Process Size | 12 nm | 3 nm |
| Transistors | 18.6 billion | Unknown |
| Die Size | 754 mm² | Unknown |
| Board Design | ||
| TDP | 150W | 140W |
| Suggested PSU | 450 W | 300 W |
| Outputs | No outputs | 1x HDMI |
| Power Connectors | 1x 6-pin + 1x 8-pin | None |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | N/A |
| CUDA | 7.5 | 12.1 |
| Shader Model | 6.8 | N/A |